Abstract
While recent studies find a strong association between the share of women in elected office and lower levels of corruption, we know less about if women in executive office cause reductions in corruption levels, and if such effects last over time. This study suggests that women mayors reduce corruption levels, but that the beneficial effect may be weakened over time. Using both regression discontinuity and first difference designs with newly collected data on French municipal elections combined with corruption risk data on close to all municipal contracts awarded between 2005 and 2016, we show that women mayors reduce corruption risks. However, newly elected women mayors drive the results, while gender differences are negligible in municipalities where women mayors are re elected. Our results can be interpreted as providing support for marginalization theories, but also suggest that the women that adapt to corrupt networks survive in office.